Fasting for haemostasis in children with gastrointestinal bleeding

Review question

To determine the effects and safety of fasting for haemostasis in children with gastrointestinal bleeding.

Background

Gastrointestinal bleeding refers to loss of blood from any site of the digestive tract. If the bleeding site is near the ligament of Treitz (the muscle extending from the junction of the duodenum and jejunum to the diaphragm), it is described as upper gastrointestinal bleeding. If the bleeding site is away from the ligament of Treitz, it is described as lower gastrointestinal bleeding. People with a small amount of blood loss may have no symptoms, and have bloody vomiting or see blood in the stool with increased bleeding. Chronic mild gastrointestinal bleeding can lead to anaemia, causing fatigue or pallor; acute massive gastrointestinal bleeding can be life-threatening.

Gastrointestinal bleeding has two scenarios in children. The first is a complaint of attending the emergency department as a symptom of diseases such as ulcers, gastritis, Mallory-Weiss tears, anorectal fissures, allergic colitis, infectious colitis, intussusception, Henoch-Schonlein purpura, and Meckel's diverticulum. The second occurs in critically ill children hospitalised in intensive care units and is caused by stress-induced gastropathy. The incidence of the first scenario has not been well established, but there are substantial cases in paediatric intensive care units in the second scenario. Whatever the cause of gastrointestinal bleeding, fasting is believed to be necessary due to the fear that eating may affect haemostasis or aggravate bleeding.

Study characteristics

We searched EBM Reviews- Cochrane Central Register of Controlled Trials (CENTRAL) (May 2016), Ovid MEDLINE(R) (1946 to 3 May 2016), EMBASE (1980 to 2016 Week 18), Chinese Biomedical Database (CBM) (1978 to 3 May 2016), China National Knowledge Infrastructure (CNKI) (1979 to 3 May 2016), VIP Database (1989 to 4 May 2016) and Wanfang Data (1990 to 4 May 2016). We used no restrictions on language or study setting and limited searches in CNKI and Wanfang Data to medical field.

We did not find any randomised controlled trials or quasi-randomised controlled trials providing information about the effects and safety of fasting for haemostasis in children with gastrointestinal bleeding.

Key results

A randomised study is required to compare the effects and safety of fasting with feeding for haemostasis in children with gastrointestinal bleeding, especially for critically ill children, who are at high risk for gastrointestinal bleeding.

Authors' conclusions: 

There is currently no information available from RCTs or quasi-RCTs to support or refute the use of fasting for haemostasis in children with gastrointestinal bleeding.

Read the full abstract...
Background: 

Gastrointestinal bleeding refers to loss of blood from any site of the digestive tract. In paediatric clinical practice, it is usually a complaint of children attending the emergency department as a symptom of diseases such as ulcers, gastric or oesophageal varices, gastritis, Mallory-Weiss tears, anorectal fissures, allergic colitis, infectious colitis, intussusception, Henoch-Schonlein purpura, and Meckel's diverticulum; it also occurs with high incidence in critically ill children hospitalised in intensive care units and is caused by stress-induced gastropathy. No matter what the cause of gastrointestinal bleeding, fasting is believed to be necessary due to the fear that eating may affect haemostasis or aggravate bleeding.

Objectives: 

To assess the effects and safety of fasting for haemostasis in gastrointestinal bleeding in children.

Search strategy: 

We searched EBM Reviews - the Cochrane Central Register of Controlled Trials (CENTRAL) (May 2016), Ovid MEDLINE(R) (1946 to 3 May 2016), EMBASE (1980 to 2016 Week 18), Chinese Biomedical Database (CBM) (1978 to 3 May 2016), China National Knowledge Infrastructure (CNKI) (1979 to 3 May 2016), VIP Database (1989 to 4 May 2016) and Wanfang Data (1990 to 4 May 2016). We used no restrictions on language or study setting and limited searches in CNKI and Wanfang Data to the medical field.

Selection criteria: 

Randomised controlled trials (RCTs) or quasi-RCTs in children with gastrointestinal bleeding that compared fasting with feeding.

Data collection and analysis: 

Two review authors independently screened the literature search results, and there were no disagreements.

Main results: 

We identified no RCTs or quasi-RCTs that compared the effects and safety of fasting with feeding for haemostasis in children with gastrointestinal bleeding. No study fulfilled the criteria for considering studies for our review.